native-agent-evals SKILL.md skillA

native-agent-evals from kbediako/evergreen-codex-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add kbediako/evergreen-codex-skills/native-agent-evals@git:20260912.7ff2e86

Pins git:20260912.7ff2e86 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
kbediako/evergreen-codex-skills
Path
skills/native-agent-evals/SKILL.md
Versions recorded
4
Latest version
git:20260912.7ff2e86, committed Sep 12, 2026
First version recorded
Jul 26, 2026
Size
2,578 bytes
Repository stars
5
License
MIT

Test native Codex agent behavior or audit rollout evidence for lifecycle, tool use, and skill invocation. Use for agent-runtime claims, not ordinary task validation.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Native agent evals

Define one observable claim with pass, fail, and inconclusive outcomes. Choose a small probe that can falsify it. Keep production mutations outside the probe unless explicitly requested.

## Run a probe

Before running native agents, load [native-subagents-first](../native-subagents-first/SKILL.md) for the lifecycle and worker contract. Exercise only the behavior under test: tool availability, completion, waiting, fanout, interruption, resume, or skill invocation. For example, fanout needs independent scopes; a lifecycle check may need only one harmless child.

For comparisons, hold model, effort, role, tools, prompt, permissions, and artifact state fixed. Change only the named treatment. Mark mismatched conditions as confounded. Repeat independent behavior comparisons before promoting a claim.

Consume the complete child answer and verify its relevant evidence. Lifecycle status alone does not prove that the requested task succeeded.

## Audit or export evidence
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add kbediako/evergreen-codex-skills/native-agent-evals@sha256:48e18902d93649c0

The line at the top of this page pins the label git:20260912.7ff2e86, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_p4iop4423umgof3s.svg)](https://markdownregistry.com/a/art_p4iop4423umgof3s)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 4 versions of this file, between Jul 26, 2026 and Sep 12, 2026, about one every 16 days.

versioncommittedcommitsizeaudit
git:20260912.7ff2e86 latest2026-09-12 7ff2e86 2,578 BA view · diff
git:20260904.11aaa2e2026-09-04 11aaa2e 3,166 BA view · diff
git:20260727.5fd1ef72026-07-27 5fd1ef7 3,164 BA view · diff
git:20260726.0ca58582026-07-26 0ca5858 2,738 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (2578 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

kbediako/evergreen-codex-skills · 5 stars · license MIT · pushed 2026-10-16 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_p4iop4423umgof3s
GET https://markdownregistry.com/api/v1/resolve?ref=kbediako/evergreen-codex-skills/native-agent-evals
GET https://markdownregistry.com/api/v1/blob/48e18902d93649c06e089ddb74366747afed94995e167d91feddc5dc4ff274f2

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from kbediako/evergreen-codex-skills

challenge-assumptions SKILL.md skill
kbediako/evergreen-codex-skills · skills/challenge-assumptions/SKILL.md · Investigate uncertain or disputed causes, failed repairs, and high-impact diagnostic decisions. Skip straightforward…
git:20260912.7ff2e86 · audit A · 5 stars
consult-chatgpt-pro SKILL.md skill
kbediako/evergreen-codex-skills · skills/consult-chatgpt-pro/SKILL.md · Consult ChatGPT Pro when requested or when an unresolved, consequential question needs an external reasoning review…
git:20261016.f5f091d · audit A · 5 stars
fast-iteration SKILL.md skill
kbediako/evergreen-codex-skills · skills/fast-iteration/SKILL.md · Use throughout software engineering work, including planning, implementation, debugging, refactoring, testing, and…
git:20260929.5b5c437 · audit A · 5 stars
long-poll-wait SKILL.md skill
kbediako/evergreen-codex-skills · skills/long-poll-wait/SKILL.md · Monitor an external process, CI run, or cloud job until an authoritative terminal result or explicit handoff. Excludes…
git:20260912.7ff2e86 · audit A · 5 stars
native-agent-deliberation SKILL.md skill
kbediako/evergreen-codex-skills · skills/native-agent-deliberation/SKILL.md · Compare competing strategies through independent agent viewpoints when a consequential decision needs evidence…
git:20260912.7ff2e86 · audit A · 5 stars
native-agent-skill-validation SKILL.md skill
kbediako/evergreen-codex-skills · skills/native-agent-skill-validation/SKILL.md · Test whether a Codex skill triggers appropriately or changes behavior using fresh controlled agent trials. Use for…
git:20260912.7ff2e86 · audit A · 5 stars
native-codex-coordination SKILL.md skill
kbediako/evergreen-codex-skills · skills/native-codex-coordination/SKILL.md · Choose between solo work, native-agent execution, deliberation, evaluation, and waiting when the appropriate…
git:20260912.7ff2e86 · audit A · 5 stars
native-subagents-first SKILL.md skill
kbediako/evergreen-codex-skills · skills/native-subagents-first/SKILL.md · Coordinate native Codex subagents for requested parallel work or independent execution and review scopes. Use for…
git:20260912.7ff2e86 · audit A · 5 stars
voice-delegate-guidance SKILL.md skill
kbediako/evergreen-codex-skills · skills/voice-delegate-guidance/SKILL.md · Coordinate operative Codex Voice handoffs through delegated work, including transcript-tail delivery, or honor explicit…
git:20260912.7ff2e86 · audit A · 5 stars
wait-for-subagents-patiently SKILL.md skill
kbediako/evergreen-codex-skills · skills/wait-for-subagents-patiently/SKILL.md · Wait for quiet native Codex children and reconcile their state before declaring a stall, interrupting, replacing, or…
git:20260912.7ff2e86 · audit A · 5 stars

Every file in kbediako/evergreen-codex-skills

Browse by kind, by grade A, or by owner.